Stereochemical studies on medicinal agents. 25. Absolute

Jun 1, 1981 - (Baker), 1 L of Triton X-100 (NEN), 169 of Omnifluor (NEN)] and measured in a Searle Mark liquid scintillationcounter (45% efficiency). ...
0 downloads 0 Views 562KB Size
J. Med. Chem. 1982,25,9-12 in ice-cold 0.05 M sodium-potassium phosphate buffer (pH 7.4). After centrifugation at 48000g for 20 min (Sorvall RC2-B), the supernatant was discarded, the pellet was resuspended in distilled water, and the process was repeated. The final pellet was resuspended in 50 volumes of ice-cold 0.05 M Tris buffer (pH 7.4). Aliquots of brain tissue (final concentration 1.25 mg/mL), [3H]spiperone(100pM), and drugs (freshly prepared in distilled water) were incubated in buffer (final assay volume 2 mL) for 30 min at 37 “C. The binding reaction was terminated by fitration in vacuo over Whatman GF/B filters, and radioactivity was extracted overnight in 6 mL of scintillation fluid [ l L of toluene (Baker), 1 L of Triton X-100 (NEN), 169 of Omnifluor (NEN)] and measwed in a Searle Mark 11liauid scintillation counter (45% efficiency). Specific [aH]spipero& binding was defined as the

9

difference between binding in the absence and in the presence of 1pM (+)-butaclamol. The negative logarithm of the concentration of drug producing a 50% inhibition of specific binding (pICM)values were estimated graphically from logarithmic Hill plots and converted to ICMvalues.

Acknowledgment. We gratefully acknowledge the hospitality given by H. I. Yamamura to one of us (C.J.G.) and thank Marla Bliss for technical assistance in performing the [3H] spiperone binding. Research performed at the University of Arizona was supported, in part, by Grant MH 31184 from the United States Public Health Service.

Stereochemical Studies on Medicinal Agents. 25.l Absolute Configuration and Analgetic Potency of ~-1,2-Dimethyl-4-phenyl-4-(propionyloxy)piperidine Enantiomers David S. Fries,’ Richard P. Dodge,t Hakon Hope,$ and Philip S. Portoghese* Section of Medicinal and Biological Chemistry and Department of Chemistry, University of the Pacific, Stockton, California 95211, Department of Chemistry, University of California, Davis, California 95616, and Department of Medicinal Chemistry, College of Pharmacy, University of Minnesota, Minneapolis, Minnesota 55455. Received June 30, 1981

Enantiomers of ~-l,2-dimethyl-4-phenyl-4-(propionyloxy)piperidine (4) were employed as probes to demonstrate that opioid receptors are capable of distinguishing between the enantiotopic edges (the Ogston effect) of the piperidine ring. These enantiomers, (-)- and (+)-4eHC1, were prepared by esterification of the corresponding alcohols, (+)and (-)-4a. Single crystal X-ray studies of (-)-4a.HCl reveal that it possesses the 2R,4S absolute configuration. Analgetic testing in mice (hot-plate) and receptor binding studies indicate that (-)-(2S94R)-4.HC1is approximately ten times more potent than its enantiomer. The results are consistent with the operation of the Ogston effect in the interaction of achiral4-phenylpiperidineswith opioid receptors. Additionally, it is suggested that the piperidine ring of these and other closely related 4-phenylpiperidines bind within a receptor subsite cleft whose dimensions ligands. exclude diequatorial 2,6- and 3,5-dimethyl-substituted

It is now generally recognized that differentiation of chemically like, paired groups (enantiotopic groups) in a substrate of the Caabc type (1) can be displayed by en-

7

Pro-4S

I

a-C-a Pro-4R

I

OH

1

2

EtOOC

3

zymes.2 Thus, substrates which contain a plane of symmetry often are found to undergo enzymatic transformation specifically on one of their enantiotopic groups. In his now classical paper, Ogston was the first to point this out in connection with the enzymatic conversion of citrate (2) in the Krebs ~ y c l e . ~ ?This ~ type of selectivity is sometimes referred to as the “Ogston e f f e ~ t ” . Although ~ the Ogston effect can be demonstrated unequivocally in enzyme-catalyzed reactions through radiolabeling experiments, a similar approach cannot be employed to investigate the interaction of ligands with noncatalytic recognition sites. In an effort to investigate the Ogston effect in the action of narcotic analgetics, we had employed a methyl group t o label the enantiotopic edges of the piperidine ring in Section of Medicinal and Biological Chemistry, University of the Pacific. Department of Chemistry, University of the Pacific. 8 University of California, Davis. * Address correspondence to University of Minnesota.

*

0022-2623/82/1825-0009$01.25/0

3.6 The methyl group was selected because it is known that the C-5 and C-6 methyl groups interfere with ligand-receptor association of the less potent enantiomers of methadone and isomethadone, respectively?v8 Initially, the C-3 position was the site of attachment of the methyl group to the pro-4R and pro-4S enantiotopic edges of 3 in these studies.6 Later publications discussed related compounds substituted at other centers.“14 However, interpretation of the results was complicated by the finding that alkyl substitution vicinal to the C-4 center induced a chiral orientation of the phenyl (Ph) group. In fact, the feature common to all of the analgetically more potent enantiomers is the sign of the torsion angles between the Ph and piperidine ring.7 (1) For paper 24,see Poupaert, J. H.; Portoghese, P. S. Bull. SOC. Chim. Belg. 1977,86,863. (2) Hanson, K.R.Annu. Rev. Biochern. 1976,45,307. (3) Ogston, A. G.Nature (London) 1948,162,963. (4) Bentley, R.Nature (London) 1978,276,673.

(5) Alworth, W. L. “Stereochemistry and Its Application in Biochemistry”; Wiley-Interscience: New York, 1972. (6) Larson, D.L.; Portoghese, P. S. J. Med. Chem. 1973,16,195. (7) Portoghese, P. S. Acc. Chem. Res. 1978,11, 21. (8) Portoghese, P. S.Annu. Rev. Phurmacol. 1970,10, 51. (9) Portoghese, P.S.;Gomma, Z. S.; Larson, D. L.; Shefter, E. J. Med. Chem. 1973,16,199. (10) Bell, K. H.;Portoghese, P. S. J. Med. Chem. 1973,16,589. (11) Bell, K.H.; Portoghese, P. S.; J. Med. Chem. 1974,17, 129. (12) Fries, D.S.;Portoghese, P. S. J. Med. Chem. 1974,17, 990. (13) Shefter, E.;Portoghese, P. S. J. Med. Chem. 1976, 19, 55. (14) Fries, D.S.; Portoghese, P. S. J. Med. Chem. 1976,19, 1155. 0 1981 American Chemical Society

Fries et al.

10 Journal of Medicinal Chemistry, 1982, Vol. 25, No. 1

Table I. Analgetic Potency and Receptor Binding of @-1,2-Dimethyl-4-phenyl-4-( propiony1oxy)piperidine Enantiomers compda confign ED,,,b mg/kg ICso,dfiM (-)-4 2S,4R 0 . 5 3 (0.40-0.71)c 1.2 2R,4S 5.5 (3.8-7.8) 11.0 ( + 1-4 0.52 (0.36-0.74) 1.3 (1.1-1.5)e 1.2

(r1-4 3e

2.0

morphine Tested as HCI salts of 4 in aqueous solution. The sign rotation is that of the salt. Tested by the hot-plate procedure after sc injection into mice.I6 Confidence interval (95%). The concentration of test agent required to displace 50% of stereospecifically bound [ 3H]etorphine (2.5 nM) from opioid binding sites in guinea pig brain homogenate. Each value is the mean of three determinations. e Reference 6. a

Figure 1. The absolute configuration of (-)-4a-HC1 as determined from single crystal X-ray studies.

In this paper we describe our effort to sort out conformational factors from configurational effects through the use of (+)-4 and (-)-4 as receptor probes. These enanMe

I

I

I

RO

: OEt

RQ 2R, 4s

2S,4R 4, R - C Q C ~ H S 4a, R : H

tiomers were employed because the 2-Me substituent is not expected to induce rotational effects on the equatorial 4-Ph group. Consequently, the receptor enantioselectivity of 4 might reflect the ability of the 2-Me substituent in one of the enantiomers to interfere with ligand-receptor association. Chemistry. The preparation of (f)-/3-1,2-dimethyl-4phenyl-4-piperidinol [(f)-4a] was accomplished using the procedure of McErlane and Casy.16 Optical resolution of the racemate was carried out by fractional crystallization of the diastereomeric salts formed with an equivalent amount of (+)- or (-)-di-p-toluoyltartaric acid. Upon treatment with ",OH, these salts afforded the corresponding bases, (-)-4a and (+)-4a, respectively. Treatment of each of these enantiomeric alcohols with propionyl chloride in dry toluene produced the corresponding desired esters, (+)-4+HC1and (-)-4.HC1, which were employed in the biological testing. Single crystal X-ray studies were performed on the alcohol, (-)-4a.HCl, derived from its base, (-)-4a, in order to determine the absolute configuration. The absolute stereochemistry of (-)-4a.HC1 was found to be 2R,4S (Figure 1). This also is the configuration of its free base, (-)-4a, which was converted to the ester (+)-4-HC1. Based on the reasonable assumption that the stereochemical integrity of the chiral centers remain intact during the esterification reaction, the ester derived from (-)-4a is designated as (+)-(2R,4S)-4*HCland its enantiomer as (-)(2S,4R)-4oHCl. Pharmacology. The racemic and enantiomeric forms of 4 were evaluated using the hot-plate procedure16after sc administration in mice (Table I). Inhibition of the binding of [3H]etorphine to putative opioid receptors in guinea pig brain homogenate was performed by the method of Pert and Snyder17as modified by Pasternak et a1.18 It

EtOCO B

A

Figure 2. Projection formulas of two Ph rotamen of (-)-(2S,4R)-4.

can be noted that (4-4 is twice as potent as morphine and ten times more potent than its enantiomer. This potency difference between enantiomers is paralleled by bind.ing studies which give an ICmratio of approximately nine. It is therefore reasonable to assume that receptor-related events are responsible for the enantioselectivity. Stereostructure-Activity Relationship. Molecular models reveal that the 2-Me substituent of 4 should have little if any influence on the phenyl (Ph) rotamer population when these groups are in the highly favored diequatorial conformation. It is therefore a reasonable assumption that each enantiomer possesses approximately equal populations of Ph rotamers A and B as illustrated by the projection formulas of (-)-4 (Figure 2). This is distinctly different from other prodines that possess alkyl substituents vicinal to the Ph group. In such cases, a gauche alkyl-phenyl interaction induces a skewed Ph rotamer distribution.' Consequently, it seems plausible that the greater affinity of (-)-4 for opioid receptors can be attributed to configurational factors rather than to a methyl-induced orientation of the Ph group. Since the less potent enantiomer (+)-4 is considerably less potent than the demethyl analoguea 3, it seems likely that this effect is due to direct intermolecular steric hindrance between the 2-Me of (+)-4 and the receptor. On the other hand, the 2-Me group in the more potent enantiomer (-)-4, being located on the p r o 4 enantiotopic edge, does not adversely affect binding. Of relevance to the present study are reports that the diequatorial analogues 5 and 6 are devoid of analgetic Ph+-"

p

h

7

"

"

'

Me C2H$00

C2H$00 5

6

a ~ t i v i t y . ' ~The ~ ~ ~fact that even high doses of 5 and 6 (15) Casy, A. F.; McErlane, K. M. J. J. Chem. Sac. 1972, 726. (16) Eddy, N. B.; Leimbach, D. J.Pharmacal. Exp. Ther. 1953,107, 385. (17) Pert, C. B.; Snyder, S. H. Science 1973, 179, 1011.

(18) Pasternak, G. W.; Wilson, H. A.; Snyder, S. H. Mol. Pharmacal. 1975, 11, 340.

Journal of Medicinal Chemistry, 1982, Vol.25,No.1 11

Substituted Piperidine Enantiomers

produce no analgesia is in contrast to the less potent enantiomer (+)-4, which is approximately as potent as meperidine. This suggests that opioid receptors contain a cleft whose dimensions exclude a piperidine ring which contains 2,6- or 3,5-diequatorial methyl groups but which is capable of accommodating a single 2-Me group on either enantiotopic edge. Accordingly, there is sufficient play in the binding so that either (+)-4 or (-)-4 can bind within the cleft, with the latter having greater affinity. Consistent with this view is the fact that the 2,6- and 3,5-dimethyl isomers (*)-721 and (*)-ti9 possess activity

I hI3

CzHSCOO

C,H&OO

7

8

comparable to the less potent enantiomer (+)-4. Presumably, these compounds are capable of binding within the cleft because the width of molecules that contain an axial-equatorial arrangement of methyl groups is less than those that contain a diequatorial arrangement (5 and 6). I t is significant that the observed potencies of various stereoisomericand regioisomeric phenylpiperidines are not simply the sum of the contribution of the individual methyl groups. This suggests that there are sufficient differences in the interaction of these ligands with opioid receptors to affect the binding contributions of individual moieties. Thus, the position of methyl substitution in the phenylpiperidine analgetics to some degree modifies their binding mode to opioid receptors. In summary, the stereostructure-activity relationship of (+)- and ( 4 - 4 is consistent with the operation of the Ogston effect in the interaction of symmetric 4-phenylpiperidines (e.g., meperidine and 3) with opioid receptors. It appears that the pro-4S enantiopic edge of the piperidine ring next to the basic nitrogen of 3, in which an equatorial Me substituent interferes with binding, is more intimately involved in the receptor interaction than the corresponding position on the pro-4R enantiotopic edge. It is suggested that the piperidine ring of these ligands binds within an opioid receptor cleft. Moreover, the width of the cleft excludes ligands beyond a certain width. Finally, the nonadditivity of analgetic potency contributed by methyl substitution on the piperidine ring of 4phenylpiperidenes is consistent with the mode of interaction of the ligand within the receptor subsite being dependent upon the location of the Me group in the molecule. Experimental Section Melting points were determined in open glass capillaries in a Thomas-Hoover Uni-Melt apparatus and are uncorrected. Elemental analyses were performed by M-H-W Laboratories, Phoenix, AZ, and agree with theoretical values within f0.4%, Optical rotations were obtained using a Perkin-Elmer 241 MC polarimeter. Liquid scintillation spectrometry was done with a Beckman CPM 100. Infrared and proton NMR spectra on each product were obtained on Perkin-Elmer 251 and JOEL Minimar 100 instruments, respectively. In all cases, spectral data were consistent with the assigned structure. Optical Resolution of (&)-@-1,2-Dimethyl-4-pheny1-4piperidinol [(*)-4a]. Racemic 4a was prepared following the methods described by Casy and McErlane.lS The fi diastereomer employed in the resolution procedure had mp 106-108 OC and (19) Nazarov, N. I.; Sorokin, 0. I. Zzu. Akad. Nauk. 1960, 872. (20) Sorokin, 0. I. Zzu. Akad. Nawk. 1960, 460. (21) Casy, A. F.; Coates, J. E.; Rostron, C. J. Pharm. Pharmacol. 1976, 28, 106.

6 (CDC13)2.3 (1-Me) and 1.07 (2-Me) in agreement with reported values.ls Racemate 4a (1.20 g, 5.58 mmol) and (-)-di-ptoluoyltartaric acid (2.28 g, 5.88 mmol) were dissolved in 40 mL of MeOH with the aid of heating and stirring. The solution was filtered, the MeOH was removed in vacuo, and the residue was dissolved in 40 mL of hot 2-butanone. The mixture was allowed to stand for 4 h at 23 'C and was filtered to afford 2.0 g of a salt: mp 155-158 'C; [.Iz3D -88.1' (c 1, MeOH). Two subsequent crystallizations from MeOH gave 1.28 g [mp 156-159 'c; [a]=D -89.3' (c 1, MeOH)] and then 0.86 g of salt [mp 158-159 'c; [cY]~D -92.0' (c 1, MeOH)]. Further recrystallizations from MeOH failed to change the [.ID or mp. Anal. (C33H37N09)C, H, N. The resolved salt (0.85 g) was dissolved in HzO (10 mL) and made basic (pH 11) with NH40H. The solution was extracted with CHC13 (4 X 2 mL), the combined extracts were dried (Na2SO4),and the solvent was removed in vacuo. The remaining oil was dissolved in 25 mL of boiling hexane and allowed to stand at room temperature in an unstoppered Erlenmeyer flask. The needle-like crystals which formed were colleded by fitration and dried to give 190 mg (35%)of (+)-4a: mp 85-87 'C; [cY]=D+34.2' (c 1, CHC13), +20.1° (c 1, MeOH). Ethereal HCl(O.2 mL) solution was added to 70 mg of (+)-4a dissolved in 10 mL of anhydrous EtzO. The solid salt which formed was collected by filtration and immediately dissolved in 8 mL of warm acetone. The flask was stoppered and left to stand at room temperature. The needle-shaped crystals which formed were isolated by filtration and dried in vacuo to yield 57 mg of (+)-4*HC1: mp 198-200 "c; [.Iz1~ +0.5' (C 1, MeOH). To ensure that isomerization had not occurred in converting (+)-4a to its HCl salt, a sample of (+)-4a.HCl was converted back to the free base and crystallized from hexane. The recovered 4a had [a]D of +20.0° (c 1, MeOH). The combined mother liquors from the above resolution were evaporated to dryness, dissolved in HzO, made basic (",OH), and extracted with 3 X 20 mL of CHClp After the extract was dried (Na2SO4)the CHC13 was removed, the oily residue was dissolved in hexane. On standing at room temperature, 0.78 g of partially resolved 4a, [a]=D -5.0' (c 1, MeOH), was obtained. This was dissolved in 30 mL of MeOH together with 1.53 g of (+)-di-p-toluoyltartaricacid. The MeOH was removed in vacuo, and the resulting oil was dissolved in hot 2-butanone. The precipitate that formed was collected by filtration, dissolved in 30 mL of hot MeOH, and then cooled to -10' for 12 h to afford 1.1 g of salt: mp 156-159 'c; [aIz3D +88.8" (c 1, MeOH). Recrystallization from 15 mL of MeOH at -10 "C yielded 0.91 g of crystallinematerid mp 158-160 'c; [aIBD +90.0° (c 1, MeOH). Two additional recrystallizations did not change the [a]D or mp. Anal. (C33H37N09) C, H, N. The free base (-)-4a was obtained by a procedure identical with that described for its enantiomer: yield 0.24 g (40%);mp 85-86 'c; [aIz3D-20.3' (c 1, MeOH). The salt, (-)-4wHCl, mp 196-198 'C, was employed in the single crystal X-ray studies. ~-1,2-Dimethyl-4-phenyl-4-(propionyloxy)piperidine Hydrochloride [(-)- and (+)-4-HCl]. Freshly distilled propionyl chloride (0.4 g) was added to 150 mg (0.74 mmol) of (+)-4a in dry toluene (10 mL) containing (100 mg) anhydrous KHC03 and maintained under Np The mixture was heated at 90 OC for 12 h and then cooled, and H20 (2 mL) was added. The organic phase was separated and extracted with aqueous NaHC03 solution (2 X 2 mL), and the toluene was removed in vacuo. Dissolution of the residue in a small volume of hexane and addition of several drops of ethereal HCl resulted in the formation of a semisolid. The solvent was decanted from this material and washed with hexane (2 X 2 mL). Dissolution of this material in boiling EtOAc (2 mL) and cooling afforded crystalline (-)-4*HCl: yield 80 mg (63%);mp 190-192 "c; [aIBD-5.6' (c 0.5, MeOH). An identical procedure using 150 mg (0.5 mmol) of (-)-4a yielded 90 mg (60%) of (+)-4.HC1: mp 192-193 "c;[(YIz3D +5.2' (c 0.5, MeOH). The racemic alcohol (*)-4a (0.6 g, 2.0 mmol) yielded 0.32 g (54%)of (A)-4,mp 192-194 'C. The racemate and enantiomers possessed identical spectral features. Anal. (Cl6HZ4C1NO2) C, H, N. X-Ray Crystallographic Studies. A crystal of (-)-4a.HCl crystallized from acetone was examined on a Syntex diffractometer. The following cell data were obtained a = 7.388(1) A, b = 12.276(2)A, c = 14.545(2)A. The space group is P212121, 2 = 4, d d = 1.22 g cm4. A total of 1240 reflections were ob-ed

J . Med. Chem. 1982,25, 12-18

12

with monochromatic Cu K a radiation using about 24 h for data collection in the w scan mode. The structure was found by classical Patterson and Fourier refinement methods, followed by full-matrix least-squares refinement computed on a Burroughs B6700. When the conventional R factor was 8.5%, examination of a difference map showed that the largest errors in the model were due to anisotropic vibration of the chloride ion and to the hydrogen atoms whose positions were obvious in this map. Full-matrix refiiement with anisotropic temperature factor for the chloride ion only, keeping the H atoms E's constant, converged with an R factor of 4.8%. This refinement also included anomalous dispersion corrections to the atomic scattering factors for all non-H atoms. To determine the absolute configuration, the positions were changed to f , 9, and and refined in the same way as before. This refinement on the mirror form dropped the R factor to 3.1 % , showing clearly that this second trial structure has the correct absolute configuration. Intense low-angle reflections were clearly affected by secondary extinction so that in these final refinements the 29 reflections having sine @/A less than 0.15 A-' were left out. The resulting structure is pictured in Figure 2 and the atomic positions are listed in Table 11. The bond angles and distances along with standard deviations are given in Tables I1 and IV (see paragraph at the end of paper concerning supplementary material). Opiate Receptor Binding. Guinea pig brain minus cerebellum homogenate was prepared as previously described.17J8 The homogenate was prepared for the binding assay as 2% wet homogenate pellet weight in 100 mM Tris-HC1 buffer without Na+ present. [3H]Etorphine (33 Ci/mmol, Amersham) was employed as the 3H-labeledopiate ligand. The homogenate (250FL),

[3H]etorphine (final concentration 2.5 X lo4 M), varying concentrations of (+)-, (-)-, or (+lb, and Tris-HC1 (to a fiial volume of 500 rL) were used in each determination. The homogenate was allowed 15 min equilibration time with the test compounds at 23 "C before the addition of the 3H-labeled ligand and then incubated at 37 "C for 30 min prior to fiitration (Whatman GF/B filters). The filters were placed in counting vials, and the [3H]etorphinewas solubilized from the membrane fragments by vortexing in 1mL of absolute EtOH and counted in 10 mL of dioxane-based (Bray's solution) scintillation cocktail. Each concentration was run in quadruplicate samples, and the mean of the nondisplaced minus the mean of the displaced with 200X excess of levorphanol was plotted as probit values vs. log concentration. The ICwvalues thus represent the ability of (-)-, (+)and (*)-4b to displace 50% of the [3H]etorphine bound at 2.5 nM concentration.

Acknowledgment. The authors are indebted to S. K. Guthrie for technical assistance, E. L. May of NIH, NIAMDD, for determination of analgetic activities, and to B. M. Cos of the Addiction Research Foundation, Palo Alto,CA, for help in setting up the opiate receptor binding assay. The research was supported in part by NIH Grant NS 05192 and a Scholarly Activity Grant made by the University of the Pacific to D.S.F. Supplementary Material Available: Tablea II-IV containing bond angles and distances along with standard deviations (2 pages). Ordering information is given on any current masthead page.

Isoxicam and Related 4-Hydroxy-N-isoxazolyl-2H-1,2-benzothiazine-3-carboxamide 1,l-Dioxides. Potent Nonsteroidal Antiinflammatory Agents Harold Zinnes, Jagadish C. Sircar,* Neil Lindo, Martin L. Schwartz, Arthur C. Fabian, John Shavel, Jr., Charles F. Kasulanis, Jerome D. Genzer, Charles Lutomski, and G. DiPasquale Warner-LambertlParke-DavisPharmaceutical Research Division, Warner-Lambert Go., Ann Arbor, Michigan 48105. Received June 1, 1981

A novel series of antiinflammatory agents, N-isoxazolyl-3-carboxamides of 4-hydroxy-W-1,2-benzothiazine1,l-dioxide, was synthesized and evaluated as antiinflammatory agents in the carrageenin-induced rat paw edema (CIRPE) assay and adjuvant-induced polyarthritis (AIP) assay. Several analogues were found to be equipotent or more potent than aspirin and phenylbutazone. Structure-activity relationships are discussed. One of the compounds, 4hydroxy-2-methyl-N-(5-methyl-3-isoxazolyl)-2H-1,2-benzothiazine 3-carboxamide 1,l-dioxide (3a; isoxicam), was found to be 3 times as potent as phenylbutazone in the CIRPE and in the therapeutic AIP assays. Isoxicam (3a) is presently undergoing phase I11 clinical trial as an antiarthritic drug.

Independent studies in our laboratories,l as well as studies by Lombardino et a1.,2 have shown that certain enolizable amides (1) have useful antiinflammatory propOH

Scheme I

0

t

R3

5

4a, R = CH,; R, = H or CH, 1, R, = H or alkyl; R, = aryl

b, R = Et; R, = H o r C H , OH

2 , R , = CH,;R, =

0 R3

3a, R , = CH,; R, = CH3

(1) H. Zinnes, N. A. Lindo, J. C. Sircar, M. L. Schwartz, and J. Shavel, Jr., J. Med. Chem., 16, 44 (1973). (2) J. G. Lombardino, E. H. Wiseman, and W. M. McLamore, J. Med. Chem., 14, 1171 (1971).

3

erties. More recently, Lombardino et alS3have reported on the amide with that replacement of the aryl group (R2)

0022-2623/82/1825-0012$01.25/00 1981 American Chemical Society